The great Brettanomyces taste test

Wednesday, 14. October 2015 - 16:15


As she passed me the wine - a 2012 Croatian blend priced at €10.00 ($11.25) ex-cellars – the winemaker watched me carefully, looking for my reaction. As I sniffed, I was aware of dark berry fruit, expensive oak – and an animal character that might have been caused by reduction, but I didn’t think so. “I like it,” I said. “But something bothers me. Is there a little Brett?”

Yes, she replied frankly, one of the blend’s components had indeed been affected by Brettanomyces, the industry’s most famous and controversial characteristic. But this had evidently not bothered the judges at the 2015 Decanter World Wine Awards who had given the wine a gold medal and a regional trophy.

I was honest when I said that I liked the wine, but the slight stable-floor note I had smelled and tasted would have stopped me from voting for that lofty a prize, if only because of my own experience as co-chairman of the International Wine Challenge. I was once given an Italian wine that was so faecal and so harshly textured that I found it undrinkable. The bottle bore a sticker proclaiming that it had won an IWC trophy three years earlier when, in the company of a team of professionals, I had happily agreed to its award.  None of us had noticed or objected to the Brett that must have lurked there, but since that tasting it had evidently developed into a monster.

What is Brett?

Nothing about Brettanomyces is straightforward. It even has another name – Dekkera – though one that’s rarely used by wine people. There are several different strains of Brett, each of which has a character of its own: some are more widely palatable than others. To complicate matters further, sensitivity to any of the strains varies from one human being to another. Even when two people perceive the same aromas, they may disagree widely over their acceptability. For one expert, a wine with detectable evidence of Brett is unpardonably faulty, while to another, the same wine will be greeted as complex and showing a sense of place. To untangle this web, it is worth looking at the facts over which there is no dispute.

Brettanomyces has the rare distinction of bringing together the worlds of beer and wine. A strain, Brettanomyces claussenii, was first identified at the Carlsberg brewery in Denmark by Niels Hjelte Claussen in 1903. Claussen was investigating a batch of sour English beer, which explains why the yeast came to be named after both him and the Greek for ‘British fungus’.  Innovative brewers realised that despite its association with faulty ale, Brett might contribute to the production of interesting beer. In 1908, a letter was sent to the editor of a Chicago publication called The Western Brewer: and Journal of the Barley, Malt and Hop Trades saying that the writer had been ‘advised’ to used the yeast to ‘improve the flavour and taste’ of their ale. The answer was that Brettanomyces was not a ‘commercial article’, but a ‘group of micro-organisms’ that can be isolated from… old English ales and that it is the ferment which gives to these beers their… taste and flavour’. The enquirer was told that 125cc of a Brett culture per barrel would be ‘sufficient’.

Ever since, brewers have taken a passionate interest in the contribution an intentional addition of Brett can make to some styles of beer, and its undesirability in others. Today, Chad Yakobson MSc, founder of the Crooked Stave Artisan Beer Project in Denver, enjoys a cult following for his Brettanomyces Project experiments into using Brett in Belgian-style beer fermentation.

Brett’s relationship with wine is less clear. The first time the two were scientifically linked was in 1940, when a researcher called M.T.J. Custers performed a first systematic study on 17 strains of Brettanomyces yeast and found one in a wine. Three-quarters of a century later, while studies have been conducted by scientists such as Pascal Chatonnet of Laboratoire Excell in Bordeaux, winemakers still have less understanding of the subject than brewers. All too often, wines with the tell-tale Band-Aid or horsey character are thought to be displaying ‘terroir’. In a 2005 Guardian interview, Chatonnet admitted that when he began to make wine in the 1960s, he thought brett was a typical character of wines from Graves. Now, he is clear that the yeast is actually the enemy of terroir because wines infected with it will all taste the same.

When discussing the smell and taste of Brett, much depends both on the precise strains of the yeast, their concentration, combination, and the style and age of wine in which they appear. One of the main yeasts, 4-ethylphenol (usually referred to as 4-EP), is associated with ‘Band-Aid’ characters, while 4-ethylguaiacol (4-EG) is more pleasantly spicy and 4-ethylcatechol (4-EC]) is the barnyardy villain of the piece. Some fruity, well-structured wines will conceal their Brett component more effectively than lighter styles – until the fruit component fades over time, allowing the Band-Aid or stable characters to emerge. Pinot Noir seems to be particularly susceptible to the barnyard character while Shiraz and Nebbiolo sometimes take on the spicy notes.

While awareness and understanding of Brett has grown in tandem with greater technical understanding, there is also no doubt that there is also more Brett around. This is explained by the trend towards picking riper grapes, higher pH levels, residual sugar, more barrel ageing, a move away from filtration and – especially among ‘natural’ winemakers - a dramatic reduction in the use of SO2. While Brett infections have been found in vineyards and on grapes, barrels are often seen as the major culprit in the spread of the infection in a winery, especially as wine is racked from one cask to another.

As Randall Grahm of the Bonny Doon winery in California told me, “a certain portion of one's barrels can well be infected and others not. It's certainly then a reasonable strategy to sterile-filter the infected barrels and not filter the others.  But it seems that it is still very tricky to have real certainty about whether your wine has Brett or not, as the testing methods seem to be not 100% reliable.”

In recent times, the argument over the acceptability and even desirability of small amounts of Brett has been enlivened by the growing trendiness of unsulphured, unfiltered, wild yeast-fermented ‘natural’ wines. Supporters of these styles are generally far more tolerant of characteristics such as volatile acidity and Brett. Alice Feiring, the New York-based author of Letting Grapes Do What Comes Naturally, has written about “earthy/savoury” Brett or “sheepiness” and says that if it does not “overwhelm the wine” it’s “terrific”.

Feiring is not alone. Plenty of well-respected palates have admired wines like Château de Beaucastel in Châteauneuf-du-Pape, Château Musar in Lebanon and Henschke’s Hill of Grace, all of which have historically – in at least some vintages – been associated with Brett.

So what do consumers think? In 2008 Chris Curtin, research manager of The Australian Wine Research Institute (AWRI) gave 104 Australians samples of wine spiked with varying levels of Brett. All the wines with Brett compounds, he reported on the AWRI website, were “significantly disliked”.

Australian consumers who are increasingly unfamiliar with the taste of TCA-tainted wines thanks to the ubiquitous use of screwcaps, are becoming increasingly unused to the flavour of Brett. A major effort by the AWRI to build awareness led to a reduction in 4-EP levels from 1,000 ppb in 2000 to less than 100 ppb in 2005. Since then, the same organisation has been the first in the world to sequence the Brett genome. The focus on cleaning up Australian wine may have had some bearing on the reaction of the casual red wine drinkers to the samples in the AWRI experiment.

A taste test

I wondered whether the result would be the same if the Brett-spiked wines were given to an international panel of experienced palates. With the help of Catavino and Vrazon, organisers of the 2014 Digital Wine Communications Conference in Montreux, along with event sponsors Nomacorc and Tim Hanni MW, we created an experiment.

We told the 300 attendees – bloggers, journalists and public relations professionals, largely drawn from Europe  –  that we wanted to explore the relationships between an individual’s personal taste and specific wines. We asked them to answer a questionnaire devised by Hanni to define the “unique combination of sensitivities and values that comprise [their] personal preferences” which  Hanni calls a vinotype. Then we gave them a set of wines to taste blind, including three samples of the same basic red Bordeaux, two of which had been spiked with Brett. The participants who had no reason to imagine that they were not tasting three different wines, were invited to declare a preference. They were also asked about which, if any of the three wines, struck them as ‘complex’, ‘fresh’, ‘fruity’,  ‘industrial’, ‘rich’ and having ‘terroir’ character.

While 91 of the tasters declined to name a favourite, fewer than a quarter of the people who did have a preference opted for the ‘clean’ sub-threshold wine. The largest segment of the 214 attendees chose the most heavily spiked wine, with almost as many picking the ‘threshold’ version.

When it came to describing the wines, significantly more of the tasters thought that the two spiked samples were ‘complex’ and ‘rich’. The obviously Brett-infected sample was recognised as being less ‘fresh’ than the others, but it was also thought to be as ‘industrial’ as the ‘clean’ sample and to show less ‘terroir’ than the others. Looking at the chart, and watching the happy wine drinkers in Paris, New York and Tokyo’s booming ‘natural’ wine bars, a wine producer might be forgiven for feeling tempted to add a judicious dose of Brett to a cuvée.

Jerry Luper, winemaker at Freemark Abbey in Napa in the 1970s, told me that Brett additions were indeed made to Napa wines in those days, though not by him. “It was when Mondavi was working with Mouton Rothschild on Opus One and everyone was focused on Bordeaux-style complexity,” he recalls, but “no one really talked about it.”

Nothing has changed. According to Grahm, having Brett in your wine is “still a social taboo, rather like letting the world know you have a sexually transmitted disease.” In any case, despite one of the world’s braver and more maverick winemakers, he sees no appeal in experimenting with Brett additions to any of his wines and doesn’t “know anyone who is doing so.”

Today, wines like Château de Beaucastel and Hill of Grace are usually described as being free of Brett. Thanks to Pascal Chatonnet, levels of Brett in Bordeaux are far lower than they were, but for those who like a little ‘funk’ or ‘sheepiness’ in their wine, examples can still be found – even among the trophy winners at major wine competitions.           

Robert Joseph